2023 ISCAP Proceedings: Abstract Presentation
UAVNetAI: An AIOps Approach for Stable Network Connectivity to Enhance Seamless Unmanned Aerial Vehicle Operations in a Metropolitan Area
City University of Seattle
City University of Seattle
Essentially, the UAV acquires real-time directives for control from the operator through the network, enabling precise execution of maneuvers, navigation, and responsiveness to mutable circumstances. Furthermore, the interchange of information between the UAV and external sources, such as GPS satellites and mapping services, relies on a network connection to form the basis for safe flight planning and navigation. Moreover, UAV procures and disseminates data in actual time, such as aerial surveys, performance data, and monitoring, for timely decision-making and analysis. Collectively construed, drones depend significantly on network connectivity with ground operators and interconnected peripherals. For that reason, it is necessary to devise effective remedies to ensure enduring and steadfast network linkages (Gupta & Vaszkun, 2015; Nawaz et al., 2021; Kim et al., 2023)
In the rapidly evolving landscape of UAV technology, maintaining stable network connectivity emerges as a fundamental prerequisite for ensuring seamless operations and safety. For that reason, In this paper, we challenge how we can make network connectivity stable during Unmanned Aerial Vehicle (UAV) operations.
Previous work emphasized the necessity and importance of stable network connectivity for UAV operations. Based on the research of predictive UAV maintenance using cloud computing and IoT (Kim et al., 2023), the study underscores the critical significance of establishing a dependable network connectivity interface linking UAVs with cloud infrastructure. An alternative inquiry elucidates that the communication infrastructure of unmanned aerial vehicles significantly impacts both service quality and operational performance (Nawaz et al., 2021). Furthermore, Gupta and Vaszkun (2015) indicate the imperative of a stable and dependable network presence for the proficient utilization of UAVs. Current solutions focus on using Mobile Ad Hoc Networks (MANETs) or Vehicular Ad Hoc Networks (VANETs) in the metropolitan area to provide network connectivity for data exchange in real time between UAV and a ground control center and seamless UAV operations (Gupta & Vaszkun, 2015). However, a significant challenge lies in the likelihood of data transmission delays for UAVs due to the low quality of network links and unserved areas in the network. Additionally, in the event of network disruptions, manual intervention becomes necessary for resolution, potentially resulting in prolonged UAV downtime.
We propose a novel UAV Network connectivity solution with AIOps, which we call UAVNetAI. AIOps employs data insights, analytics, and machine learning to automate and enhance IT operations. By analyzing substantial data, it identifies patterns, addresses current issues, and utilizes AI for optimal adjustments in real time. AIOps minimizes manual intervention, accelerates incident response, and aids in predicting and preventing future problems.
UAVNetAI offers an autonomous network management solution encompassing tasks, such as network connectivity monitoring, identification of network anomalies, and proactive real-time resolution. UAVNetAI acquires data about network connectivity, including metrics like network speed, connection failures, geographic coordinates of connectivity failures, signal strength, and other relevant variables. This network log data trains the AI model, which provides actionable directives, based on precise discernments of abnormal network anomalies. UAVNetAI autonomously recalibrates flight paths or changes network access points in response to the recommendations, thereby eliminating root causes of connectivity issues during UAV operations.
In the realm of UAVs, stable network connectivity is vital for safe and effective operations. The proposed solution, UAVNetAI, can bring the stability of network connectivity to UAVs to ensure seamless operations. UAVNetAI autonomously manages networks, collects data, analyzes anomalies, and executes the suggested actions such as adjusting flight paths or changing access points to eliminate intermittent connectivity issues.
Gupta, L., Jain, R., & Vaszkun, G. (2015). Survey of important issues in UAV communication
networks. IEEE communications surveys & tutorials, 18(2), 1123-1152.
Kim, T., Ju, A., Maeng, B., & Chung, S. (2023). A Predictive Unmanned Aerial Vehicle Maintenance
Method: Using Low-Code and Cloud-Based Data Visualization. Journal of Information Systems Applied Research, 16(2).
Nawaz, H., Ali, H. M., & Laghari, A. A. (2021). UAV communication networks issues: a review.
Archives of Computational Methods in Engineering, 28, 1349-1369.